B. J. Tins, MD; V. N. Cassar-Pullicino, MD, FRCR
Department of Diagnostic Radiology, Robert Jones & Agnes Hunt Orthopaedic Hospital, Oswestry, Shropshire, SY10 7AG, UK
Epidemiology
The incidence of SUFE varies widely between races (Loder 1996a). Absolute numbers therefore need to be interpreted with knowledge of the racial composi- tion of the examined patient population. For white children in the United States an annual incidence of 3.19 per 100000 is quoted based on articles from the early 1970s (Kelsey et al. 1970; Kelsey 1971). In comparison the relative incidence of SUFE is about 4.5 for Polynesian children and only 0.1 for Indo- Mediterranean children (Loder 1996a). In black children in the United States the relative incidence is roughly 2.2 times higher than in white children (Kelsey 1971; Loder 1996a). The relative sex distri- bution for children with SUFE varies between 90%
boys in Indo-Mediterranean children and 50% boys for Native Australian/Pacific Island children. In the United States population as a whole the average sex distribution of children with SUFE is 66% boys and 34% girls (Brown 2004). The average age at onset of SUFE also varies with race and sex. Worldwide the average age at onset is 12 years for girls and 13.5 years for boys (Loder 1996a).
In Germany the incidence of SUFE is about 1:100000, with a boys to girls ratio of 3:1, the average age at diagnosis is 12.5 years for girls and 14.5 years for boys (Harland and Krappel 2002).
There is seasonal variation in the relative inci- dence of SUFE for children living beyond 40 ° lati- tude; the incidence is increased in summer (Loder 1996; Brown 2004) and this effect is more pro- nounced in white children (Brown 2004).
Increased weight (Loder 1996a) and body mass index (BMI) (Poussa et al. 2003) are risk factors for SUFE. This can potentially cause an increased inci- dence of SUFE with the growing obesity of children and teenagers in recent years.
12.1
Introduction
Slipped upper femoral epiphysis (SUFE) is usually a disease of adolescent teenagers. Diagnostic and treat- ment options are constantly evolving, partly driven by advances in imaging and therapeutic techniques, partly by a better understanding of its pathology and partly by studies of long-term outcome of treatment methods. This chapter briefly reviews the underly- ing pathological changes and clinical presentation before describing in more detail the imaging diag- nosis, treatment options, complications and their relevance for imaging follow up.
12.2.1 Epidemiology 173 12.2.2 Pathology 174
12.3 Clinical Presentation and Differential Diagnosis 175
12.4 Clinical Classifications 178 12.5 Imaging of SUFE 178 12.5.1 Radiography 178 12.5.2 MR Imaging 184
12.5.3 Computed Tomography 184 12.5.4 Ultrasound 185
12.5.5 Nuclear Medicine 185
12.6 Treatment Strategies and Problems 185 12.6.1 Conservative Treatment 189
12.6.2 Operative Treatment 189 12.7 Treatment Philosophies 191
12.8 Imaging Assessment of Surgical Intervention and Complications of SUFE 192
12.9 Conclusion and Outlook 193 References 193
Other potential causes for SUFE are manifold.
Some authors see a connection between increased mechanical stress (weight and activity/power) and decreased mechanical strength of the cartilage growth plate at the end of puberty. Steroid hormones and STH are thought to be detrimental, but a direct correlation has not been proven. Lack of vitamin C and D, calcium and thyroid hormone are risk factors (Fig. 12.1) as well as increased parathormone levels and renal osteodystrophy (Weiner 1996; Milz et al.
2002; Schultz et al. 2002; Oppenheim et al. 2003).
Deeper acetabular depth (Loder et al. 2003) and decreased anteversion of the femoral neck (Weiner 1996; Exner et al. 2002) have also been identified as risk factors. The occurrence of contralateral SUFE is quoted as being between 20%–80% (Engelhardt 2002; Harland and Krappel 2002) with a large study quoting 37% (Loder et al. 1993a). Contralateral slips can be clinically asymptomatic is in up to 71%
of cases. The longer the time to fusion of the growth plate the larger the risk (Schultz et al. 2002).
12.2.2 Pathology
The pathological changes of SUFE can be discussed as two entities; firstly the primary changes of the physis and adjacent tissues directly involved in or responsible for the development of SUFE and sec- ondly secondary changes resulting from the slip.
Primary pathological changes of SUFE arise in the physis. The cartilage of the growth plate usu- ally fails near the metaphysis since the cartilage is mechanically weakest here. Histologically fibrillar
disintegration can be seen; whether this is cause or effect of the slip is unclear (Weiner 1996; Milz et al. 2002). Stabilising elements are the perpendicu- lar alignment of the growth plate against the forces acting on it, the additional cupping of the proximal femoral epiphysis around the metaphysis and the formation of grooves in the physis. These grooves are particularly important to counteract rotational forces. The periosteal thickening, the zona orbi- cularis, around the femoral neck also increases mechanical stability (Exner et al. 2002; Milz et al.
2002). After operative internal fixation in patients with SUFE, the residual growth plate can revert to a more normal histological appearance and further growth can take place. This suggests that mechani- cal factors are probably responsible for the growth plate abnormalities (Guzzanti et al. 2003).
Secondary pathological changes affect tissues adjacent to the physis. The blood supply to the femo- ral epiphysis is delivered largely through periosteal vessels. Tears to the periosteum compromise this supply. In slipped upper femoral epiphysis the peri- osteum does tear opposite to the direction towards which the slip occurs (Fig. 12.2). On the side of the slip the periosteum is usually preserved and ini- tially fairly loose and buckled (Fig. 12.3). With time the sleeve contracts and the metaphysis begins to remodel and form a bony spur in response to the adjacent femoral head. The periosteal sleeve has to stretch over this spur. At about 2 weeks after the slip this process has caused sufficient shortening of the periosteal sleeve to result in a tear at the side of the slip should reduction of the slipped epiphysis occur.
This can then result in further compromise of the vascular supply and possible avascular necrosis of
Fig. 12.1a,b. A boy with severe hypothyroidism. Diagnosed age 13 years. Immature skeleton. a Bilateral slipped femoral epiphyses with sclerotic and irregular metaphyses at age 13 years. b At 2 years later the slip has worsened, still immature skeleton despite thyroxine therapy. Gross remodelling of the metaphyses, gross femora vara
a b
the femoral epiphysis (Boyer et al. 1981; Arnold et al. 2002a; Exner et al. 2002; Boero et al. 2003).
12.3
Clinical Presentation and Differential Diagnosis
Clinically patients often present with hip pain but the pain can also be referred to groin, thigh or knee (Harland and Krappel 2002). Referred pain and relatively mild symptoms are responsible for delay
in diagnosis and a high index of clinical suspicion in patients with unclear knee, thigh or groin pain is necessary (Ankarath et al. 2002; Kocher et al.
2004). Some authors recommend radiographs of the hips in anterior-posterior (AP) and frog lateral projection for any adolescent with undiagnosed knee or hip pain lasting for 1 week (Ankarath et al. 2002). On examination the internal rotation is often diminished and on flexion of the hip exter- nal rotation occurs (Harland and Krappel 2002).
The differential diagnosis comprises osteonecrosis, infectious or inflammatory arthritis, tumour and trauma (Lalaji et al. 2002).
Fig. 12.2a–e. A boy aged 8 years and 7 months. a AP view of the hip joints demonstrates bilateral severe slip. Generalised osteopenia. The metaphyses impinge onto the acetabuli. b Coronally reconstructed CT confi rms the fi ndings. c Axial CT dem- onstrates bilateral severe slip and ossifi cation of the torn periosteal sleeve laterally on both sides. d The slips were pinned but the deformity not corrected. e CT demonstrates dorsal position of the screw in the right femur, ideally the screw should be central within the femoral epiphysis
b
e
c
Fig. 12.3a–k. A girl aged 11 years and 6 months. Left-sided severe slip, mainly posteriorly, therefore appearances on the AP view (a) are decep- tive, the true severity is best appreciated on the lateral view (b). Coronal STIR (c), sagittal (d) and axial (e) T2-weighted MR images confi rm the severe slip. The coronal STIR image shows the elevated peciosteum medially due to the slip (laterally the peciosteum tears). Single screw fi xation in malalignment. At 14 months later (f,g) deformity of the femoral head due to impingement on the lateral acetabulum. Tc-99m isotope bone scan demonstrates increased activity in the diffusion (h) and the bone phase (i) indicating activity of the pathological process and raising the possibility on AVN. Normal right-sided hip joint. Fur- ther follow up radiograph (j) confi rms AVN of the cranial part of the femoral head and lateral notch formation. Osteotomy was subsequently performed (k)
b
c
e a
d
k
f g
h i
j
12.4
Clinical Classifications
Several classification systems exist for the assess- ment of SUFE. The most commonly used clinical classification differentiates acute, acute-on-chronic and chronic disease. This classification is important because the choice of therapy often depends on it.
Acute disease is classed as onset of symptoms less than 3 weeks prior to presentation with acute pain, acute on chronic is defined as presentation with acute pain but onset of some pain more than 3 weeks prior to presentation, chronic disease is defined as symptom onset prior to 3 weeks and presentation without acute symptoms (Loder et al. 1993b). The differentiation between acute and chronic slips can also be based on imaging criteria; this is addressed in detail in Sect. 12.5.
Other classifications define the disease as stable, when the patient is able to weight bear with crutches and as unstable if he is not (Loder et al. 1993b).
The classification of the severity of the slip is based on the slip angle as determined by imaging methods. Slip angles of < 30 ° are classed as mild, 30 °–50° as moderate and > 50° as severe (Boyer et al. 1981). If imaging does not allow accurate angle measurement the relative displacement of the femo- ral head on the neck allows for a rough estimation;
< 1/3 displacement is seen as mild slip, 1/3–1/2 as moderate and > 1/2 as severe slip (Boyer et al. 1981;
Kallio et al. 1993).
12.5
Imaging of SUFE
The diagnosis of slipped upper femoral epiphysis is made by imaging. Imaging signs of SUFE vary with the stage and severity of disease.
The earliest sign of acute disease may be wid- ening and irregularity of the physeal plate with ill defined metaphysis and epiphysis adjacent to the growth plate and spotty or streaky radiopacities in the growth plate and deossification adjacent to it (Fig. 12.4) (Klein et al. 1951; Loder et al. 1993b;
Gekeler 2002). Deossification is frequently seen affecting the whole hip joint area (Klein et al. 1951).
In the most common posteromedial slip the normal lateral overhang of the femoral head over a tangent to the lateral neck of femur disappears with progres- sion of the slip (Fig. 12.3). With further progression the femoral head visibly tilts against the acetabu-
lum and the femoral neck but stays within the ace- tabulum. Acute slips are often unstable (Fig. 12.5) (Klein et al. 1951; Gekeler 2002). Acutely a joint effusion may be seen and this is seen as a sign of instability (Harland and Krappel 2002) or acuity of the slip (Kallio et al. 1991, 1993). Cystic change of the metaphysis may be seen in acute as well as in chronic disease (Gekeler 2002). The main dif- ference between the acute and the chronic stage is the absence of metaphyseal remodelling in the acute stage. Chronic epiphyseolysis usually demonstrates varus deformity of the femoral neck and formation of bone spurs on the medial aspect of the metaphy- sis. Sloping of the medial metaphysis is also often seen in chronic disease (Fig. 12.6) (Gekeler 2002;
Leunig et al. 2002).
A systematic approach to radiographs in sus- pected SUFE is summarized in Tables 12.1 and 12.2.
12.5.1 Radiography
12.5.1.1 Positioning
The most important imaging modality is radiog- raphy. Usually AP and specialised lateral views of both hips are taken. Obtaining radiographs of both hips simultaneously allows for comparison between the sides and is also indicated because of the high incidence of bilateral disease. For accurate angle measurements exactly defined patient positioning is a must (Engelhardt and Roesler 1987; Gekeler 2002).
In English speaking countries positioning is usually done according to Southwick’s description (Southwick 1967) which is similar to an earlier description by Klein et al. (1951). The AP view is taken with the patients’ pelvis flat on the table. The beam is centred exactly in the midline between the hips. The hips are neutral or as near to neutral as pos- sible, i.e. the patellae point straight up. For the frog lateral view the hips are placed in maximal abduc- tion and external rotation. The knees are flexed and the plantar surfaces of the feet face each other. The lateral parts of the feet rest on the table. In cases where frog lateral views are not possible, true lateral radiographs can be used (examined hip extended, opposite hip flexed) (Southwick 1967).
In German speaking countries positioning is usu-
ally done according to Imhäuser. Similar to South-
wick, Imhäuser suggests exact neutral position of the legs for the AP view, the patellae point forward.
However, in cases with fixed external rotation in the hip joint he suggests to correct the external rotation by elevating the pelvis. In these cases the hip joints can not be imaged together. For the lateral view Imhäuser suggests to follow Lauenstein’s descrip- tion: the hip joint is flexed by 90 ° and abducted by 45 °. This is easily achieved with a dedicated wedge.
The calf has to be parallel to the long axis of the table to avoid rotation (Gekeler 2002).
There is an important caveat for performing lat- eral views. In high grade slips forced flexion and abduction of the hip joint can cause worsening of the slip. Standard projections should not be forced if painful for the patient (Gekeler 2002).
12.5.1.2
Radiographic Signs of SUFE
Early disease presents with ill defined meta- and epiphyses adjacent to the growth plate and irregu- lar radiopacities in and around the widened growth plate (Fig. 12.4). At this stage there is not always an actual slip seen yet. With an epiphyseal slip an axis perpendicular to the base of the epiphysis is no longer parallel to the femoral neck (Fig. 12.7). Devia- tion of > 2 ° is seen as abnormal (Freyschmidt et al.
2001). When the epiphysis begins to slide and tilt, radiographically the growth plate seems to narrow, this is partly a projectional phenomenon and partly true narrowing. In cases of the most common type of slip, the dorsomedial slip, a tangent to the lat-
Fig. 12.4a–d. A female aged 9 years 10 months. Mild slip right hip. On the AP view (a) only irregularity of the growth plate seen. Tangent to the neck of femur still intersecting the femoral head. On the lateral view (b) mild posterior slip. Ultrasound of the right hip (c) demonstrates minor dis- placement of femoral epiphysis versus the metaphysis by 2.5 mm, ultrasound of the left hip is normal (d)
a
c d
a
b
c
d
e f
Fig. 12.5a–k. A girl aged 12 years and 6 months. Initial radiographs (a) show osteopenia of the right hip and mild to moderate slip. Five days later (b) the slip has worsened indicating instability. Metaphy- seal remodelling indicates chronicity. STIR image (c) shows marked bone marrow oedema and metaphyseal impingement onto the acetabulum. Post pinning the screw is seen poorly positioned (d) with the screw too dorsal. CT (e) confi rms this, this is even better appreciated after screw removal (f). The screw was repositioned, radiographs (g,h) demonstrate the screw centrally within the femo- ral head on both projections indicating good position. This is con- fi rmed with CT (i). Nine months later the screw is surrounded by a radiolucent halo indicating loosening (j). Two years later (k) and after closure of the growth plate the screw is fi rmly incorporated again
g h
i j
k
a
b
c d
e
f
Table 12.2. Checklist for reporting of slipped upper femoral epiphysis
x AP and frog lateral views performed and of diagnostic quality?
x Slip present?
x Approximate degree of slip?
x Slip acute, acute on chronic, chronic?
x Secondary complications (chondrolysis, impingement acetabulum/metaphysis)?
Table 12.1. Radiographic signs of slipped upper femoral epi- physis
x Growth plate ill defined, possibly widened
x Base of epiphysis no longer perpendicular to the neck of femur (NOF) axis
x In typical dorsomedial slip lateral tangent to NOF no longer intersects with femoral head
x In chronic slip remodelling of metaphysis, possibly with varus (or rarely valgus) deformity of NOF
x Osteopenia
Fig. 12.6a–h. A boy aged 16 years and 6 months. AP view of both hips (a), coned AP (b) and lateral (c) view of the left hip show a severe chronic posterior slip with metaphyseal remodelling. Axial STIR (d) shows marked oedema in the neck of femur and in the growth plate. Gadolinium contrast enhanced T1-fat saturated (T1-FS) image (e) shows contrast uptake in the areas of bone marrow oedema. CT shows beginning ossifi cation of the physis and a cyst in the metaphysis (f). AP (g) and frog lateral (h) views after internal fi xation show the persistent marked malalignment of the femoral head with the neck of femur
Fig. 12.7. Alignment of the upper femoral epiphysis on the lateral view. Normally a line perpendicular to the base line of the epiphysis is parallel to the femoral neck axis. In SUFE the perpendicular to the epiphyseal base line is no longer parallel to the femoral neck axis. In chronic SUFE (not shown) the femoral neck often remodels and the femur shaft axis must then be used as reference line
eral femoral neck no longer intersects the lateral aspect of the femoral head (Fig. 12.3). The lateral view is in most cases more sensitive for the recogni- tion of SUFE (Fig. 12.4) (Klein et al. 1951; Gekeler 2002).
In chronic cases signs of metaphyseal remod- elling are seen. This comprises formation of a metaphyseal bony spur adjacent to the slip, usually dorsomedially, varus deformity of the femoral neck and possibly signs of secondary joint damage in
higher grade slips due to impingement of the bared metaphysis onto the acetabulum (Figs. 12.3, 12.5, 12.6) (Gekeler 2002; Leunig et al. 2002).
In rare cases evidence of loss of cartilage thick- ness (chondrolysis) is seen on the initial radiographs.
Chondrolysis is defined as loss of cartilage thickness
of ≥ 2 mm compared to the contralateral normal hip
or an absolute cartilage thickness of ≤ 3 mm in cases
of bilaterally abnormal hips. Normal hips have a
cartilage thickness of 4–5 mm (Loder et al. 1993b).
12.5.1.3
Radiographic Measurements
Angles measured in the AP and lateral projection are not equal to the actual slip angle. For low slip angles, the angles measured in the lateral projection are similar to the real angles (Fig. 12.7), for higher slip angles the real angles can be determined by using conversion tables. Exact positioning is prerequisite for this (Gekeler 2002). In patients with chronic slip and metaphyseal change the angle needs to be measured against the femoral shaft to avoid inac- curacies due to varus, or rarely valgus, deformity of the femoral neck (Engelhardt and Roesler 1987).
Despite optimal positioning, angle measurements can have inaccuracies of up to 5 ° and sometimes even more, especially when the hip joint is in flexion contraction (Engelhardt and Roesler 1987).
In cases where only a single AP view has been acquired there is another method to determine the slip angle. In the AP view the base of the slipped epiphysis appears elliptic. By measuring the long and the short axis of the ellipse and referring to a conversion table the slip angle can be estimated. This method is however inaccurate (Gekeler 2002).
Another classification system for the assessment of poor radiographs, where angle measurement is not possible, uses the relative displacement of the femoral head on the neck for classification. Dis- placement of < 1/3 is seen as mild, 1/3–1/2 as moder- ate and > 1/2 as severe slip (Boyer et al. 1981).
12.5.2 MR Imaging
Currently MR imaging for established or suspected cases of SUFE is used as a problem solving tool.
It is not usually the primary imaging examination of choice, this is still radiography. MR imaging is useful in equivocal cases and also allows very accurate angle measurements by choosing the most suitable imaging plane, usually the sagittal plane (Daschner et al. 1990; Harland and Krappel 2002). A few authors see MR imaging as secondary to computed tomography (CT) (Engelhardt and Roesler 1987; Engelhardt 2002), but most authors prefer MR imaging to CT in most cases (Daschner et al. 1990; Harland and Krappel 2002).
The ability of MR imaging to detect cases of SUFE early and even “preslip” is quoted repeatedly in the literature (Daschner et al. 1990; Stabler et al. 1992; Harland and Krappel 2002; Lalaji et
al. 2002). However, it is the opinion of the authors of this chapter that there is no convincing case in the literature proving that MR imaging can diag- nose SUFE in the preslip stage. Examples labelled as preslip actually demonstrate evidence of slip on the radiographs illustrated in these articles.
On MR imaging SUFE presents with oedema like signal in the growth plate and adjacent epiphysis and metaphysis, widening and loss of sharpness of the growth plate and slip of the epiphysis (Daschner et al. 1990; Stabler et al. 1992; Umans et al. 1998;
Harland and Krappel 2002; Lalaji et al. 2002).
Metaphyseal sclerosis and spur formation is seen (Figs. 12.3, 12.5, 12.6) (Umans et al. 1998). Joint effu- sion is frequently seen, this is however unspecific (Engelhardt 2002; Harland and Krappel 2002).
In the institution of the authors of this chapter, MR imaging is currently only used as a problem solving tool, particularly for cases of doubtful SUFE, including queried contralateral disease in estab- lished cases of SUFE.
The absence of ionising radiation is an advantage of MR imaging but the ease, speed and good sen- sitivity and specificity of radiographs still favour these as first line examination.
12.5.3
Computed Tomography
Computed tomography attracted early inter- est, before MR imaging became widely available (Engelhardt and Roesler 1987). Compared to radiography the radiation dose of CT is higher and for routine cases of suspected SUFE CT is not neces- sary (Harland and Krappel 2002). It might have a role for problem solving in patients who can not undergo MR imaging. It is also useful for evaluation whether closure of a growth plate has commenced, this can be important when prophylactic pinning of the asymptomatic side in patients with unilateral SUFE is contemplated (Engelhardt 2002).
In SUFE CT demonstrates physeal widening, metaphyseal irregularities with sclerosis and scal- loping and in chronic cases metaphyseal beaking (Figs. 12.2, 12.6) (Umans et al. 1998).
CT, especially with reformatting, can be useful to
determine the slip angle in difficult cases, such as
in flexion deformity in the hip joint. In these cases
it can also depict the joint alignment and possible
impingement. It can also be used to determine the
position of orthopaedic implants and in particular
protrusion of implants into the joint; this can be
ing radiation and the position of greatest slip can be determined sonographically. This theoretically avoids the problem of projectional error for the slip angle determination. Ultrasound also allows direct assessment of joint effusions (Kallio et al. 1991, 1993). However ultrasound is operator dependent and the information about bone changes and overall joint appearance is limited. Radiographs are familiar to radiologist and surgeon and ultrasound is not useful for postoperative follow up. Ultrasound has however been proposed for the follow up of asymptomatic con- tralateral hips after SUFE when a contralateral slip is queried (Castriota-Scanderbeg and Orsi 1993). In these cases it provides a sensitive screening test with- out ionising radiation that is easily complemented by radiography in positive cases.
Generally the use of a 5 MHz probe is advised. The transducer is aligned parallel to the femoral neck and circumferential scanning is performed. This allows to identify the area of maximal displacement of the epiphysis. The displacement of the epiphysis from the metaphysis can then be measured. Similarly the distance of the metaphysis from the acetabulum can be determined.
In acute SUFE an offset between meta- and epiphy- sis is seen. In typical dorsomedial slips this is easily imaged. Comparison with the contralateral side is also easy (Figs. 12.4, 12.9). In acute cases a joint effusion is usually seen. Joint effusion is diagnosed if there is
> 2 mm difference to the normal side or a joint collec- tion of > 6 mm on the symptomatic side. Ultrasound imaging of chronic slips is more difficult, as with time remodelling takes place which causes narrow- ing of the physeal step. The size of the physeal step is inversely correlated to the duration of symptoms and directly correlated to the degree of displacement (Kallio et al. 1991). However, small physeal steps can also occur in delayed closure of a physis and Perthes disease (Harland and Krappel 2002).
12.5.5
Nuclear Medicine
Tc-99m labelled scintigraphy of the hip joints with pin hole collimation technique has been used to assesses the status of femoral head, physis and the hip joint as a whole in patients with SUFE. Bone scin- tigraphy was (and nowadays rarely is) used for spe- cific management decision during treatment, after the primary diagnosis was made radiographically.
Bone scintigraphy can help diagnose avascular necrosis (AVN) of the femoral head by demonstrat- ing initially absent uptake and later reperfusion and remodelling with increased uptake. It can also assess the status of the growth plate. Radiographically it is often difficult to decide the extent of physeal closure and a Tc isotope bone scan is of help here, increased uptake indicates persistence of the growth plate (Figs. 12.3, 12.10) (Smergel et al. 1987).
Bone scintigraphy can also be used for diagnos- ing chondrolysis after or with SUFE. Generalised increased uptake around the affected joint, prema- ture closure of the femoral capital physis and the apophysis of the greater trochanter with decreased uptake in theses areas are suggestive of chondrolysis (Mandell et al. 1992; Warner et al. 1996).
To the authors’ knowledge, nuclear medicine is no longer in use for diagnosis or management of SUFE and has been supplanted by CT or MR imaging.
12.6
Treatment Strategies and Problems
There is a large variety of treatment strategies and
techniques in orthopaedic textbooks and journals
and a number of quite contrasting treatment phi-
a b
c
d
e f
Fig. 12.8a–h. A boy aged 8 years and 1 month; bilateral SUFE. Intraoperatively no obvious cartilage transgression but screw tips near to the cortical bone of the femoral head (a,b right; c,d left). Axial CT images were equivocal (e,f). Coronal reconstructions (g,h) demonstrate the right screw contained within the femoral head, the left screw penetrates into the fovea centralis. This case demonstrates the need for thin slices and reconstructions for the assessment of screw position after pinning
Fig. 12.9a–d. A girl aged 11 years and 4 months. AP (a) and frog lateral (b) show a moderate slip of the left upper femoral epiphysis. Ultrasound of the left hip (c) demonstrates displacement by 1.3 cm, the right hip (d) demonstrates normal alignment of femoral epiphysis and metaphysis
a b
c d
Fig. 12.10a–f. A boy aged 11 years and 4 months. Severe left sided slip (a). The diffusion phase of an isotope bone scan (b) demonstrates marked increased uptake around the epiphysis of the left proximal femur, no evidence of AVN. The bone phase images demonstrate the slip on the left side (c) and normal appearances on the right side (d). Single screw fi xation in malalign- ment (e) leads to formation of a lateral notch of the femoral head due to impingement (f)
a b
c d
e f
doned (Arnold et al. 2002).
12.6.2
Operative Treatment
All surgical treatment methods for SUFE aim to stabilise the slip. The second goal is to achieve bio- mechanically sound alignment to prevent or lessen secondary osteo-arthrosis. This can be achieved by reducing the slipped epiphysis or osteotomy (Arnold et al. 2002).
The importance of the time span from diagnosis to treatment is subject to debate. There is no clear evidence that immediate intervention in SUFE fares better than urgent (within a few days) intervention, with bed rest until surgery. Treatment should be per- formed by specialist surgeons (Loder et al. 1993b;
Arnold et al. 2002; Exner et al. 2002).
The proponents of bone grafts claim a low rate of AVN in its favour (Adamczyk et al. 2003). The implantation of two Kirschner pins or small cannu- lated screws offers the advantage of rotary stability (Exner et al. 2002), but there is evidence that the complication rate (AVN, chondrolysis) increases with the number of metal implants (Fig. 12.12) (Engelhardt 2002; Hackenbroch et al. 2002; Tok- makova et al. 2003) and many surgeons prefer either a single, larger cannulated screw or a DHS (Boyer et al. 1981; Aronson and Carlson 1992; Loder et al.
1993b; Arnold et al. 2002a-c; Ballard and Cos- grove 2002; Engelhardt 2002; Hackenbroch et al. 2002; Carney et al. 2003; Tokmakova et al. 2003).
A DHS allows for continued growth, thus potentially reducing length discrepancies between the two legs and avoiding shortened femoral necks with elevated trochanters (Arnold et al. 2002; Exner et al. 2002;
Hackenbroch et al. 2002).
Fig. 12.11a,b. A boy aged 14 years and 8 months. Right-sided moderate to severe slip (a). An intraoperative image shows cartilage injury of the femoral head by the guidewire (b). This should be avoided. Note also the good alignment of femoral
head and neck due to spontaneous reduction of the slip b
a
Fig. 12.12a–f. A girl aged 9 years and 4 months. Severe left SUFE (a,b). Pre-treatment images demonstrate impinge- ment of the metaphysis on the acetabulum. Three pins were inserted, spontaneous reduction of the slip during positioning for surgery. No initial complication (c,d). Further radiographs 4 months later (e,f) demonstrate avascular necrosis and col- lapse of the cranio-lateral part of the femoral head. Avascular necrosis is positive correlated with reduction of a slip and the number of implants. Therefore two risk factors in this case a
b
c
d
e
f
The injury of joint cartilage by the implants must be avoided because of the increased risk of chondrolysis after cartilage injury (Arnold et al.
2002; Exner et al. 2002; Leunig et al. 2002; Jofe et al. 2004). This includes the avoidance of tran-
sient cartilage injury by a guidewire during surgery
(Fig. 12.11). The use of Kirschner pins is seen criti-
cally by some authors as they can work themselves
loose with time, resulting in an increased risk of car-
tilage damage (Arnold et al. 2002).
From a simple biomechanical viewpoint reduc- tion is preferable to in situ fixation without reduc- tion. However active reduction increases the risk of AVN; as the femoral head begins to slip in one direc- tion the periosteal sleeve on the opposite side tears and with it the blood vessels contained in it. A large proportion of the blood supply to the femoral head is then maintained by vessels on the side of the slip.
At about 2 weeks after the slip, metaphyseal remod- elling with bony spur formation and tightening of the joint capsule has taken place. At 2 weeks and more after the slip, attempts at reduction therefore carry a significant risk of disruption of the surviv- ing capsular blood supply thereby causing femoral head necrosis (Boyer et al. 1981; Arnold et al. 2002;
Exner et al. 2002; Boero et al. 2003). This is the reason why some surgeons suggest open reduction for chronic or even acute SUFE, which makes it pos- sible to resect the bone spur and reduce the slipped head under direct vision, sparing the blood supply. In acute SUFE there should theoretically be no greatly increased risk of AVN from reduction (Boyer et al.
1981; Arnold et al. 2002; Exner et al. 2002).
Some surgeons use the classification of stable (patient can weight bear) and unstable (patient can not weight bear, even with crutches) for the decision whether reduction is safe (yes in unstable, no in stable) (Harland and Krappel 2002). However, many sur- geons will not try to reduce SUFE at all because of the risk of AVN (Hackenbroch et al. 2002; Carney et al.
2003; Tokmakova et al. 2003) especially in unstable SUFE (Ballard and Cosgrove 2002).
12.6.2.3 Osteotomy
For corrective osteotomy after SUFE there are four principal techniques. These are subcapital, neck
12.7
Treatment Philosophies
The previous paragraph outlined common treat- ments of SUFE, each with its own problems. The treatment choice is partly guided by different phi- losophies and partly by personal preference.
In the United States most cases of SUFE are treated with in situ surgical fixation only. No attempt of reduction is made (though reduction does often occur with closed fixation) and primary osteotomy is unusual. Most centres opt now for a single can- nulated screw for fixation (Aronson and Carlson 1992; Loder et al. 1993b; Ballard and Cosgrove 2002; Carney et al. 2003; Tokmakova et al. 2003;
Jofe et al. 2004).
In central Europe and in particular in the German speaking countries the treatment of SUFE is more aggressive. There are guidelines from orthopaedic and trauma surgeon organisations advising in situ fixation for slips < 30 °, in situ fixation plus sub- trochanteric osteotomy for slips between 30 ° and 50 °–60° and for slips of > 50°–60° open reduction and subcapital osteotomy with neck of femur short- ening is advised. Some surgeons still try to achieve active reduction before fixation. A single cannulated screw is usually preferred to two Kirschner pins or a DHS (Arnold et al. 2002a–c; Engelhardt 2002;
Exner et al. 2002; Hackenbroch et al. 2002; Har- land and Krappel 2002; Leunig et al. 2002; Schai and Exner 2002).
Whether removal of surgical fixation screws is advisable is controversial (Aronson and Carlson 1992).
Differences in philosophy can partly explain the
difference in treatment approaches. The surgical
philosophy favoured in the US aims for minimal
trauma at the time of slip to reduce the risk of imme-
diate complication. This results in an increased risk
of long term arthrosis. The approach favoured in central Europe is aggressive at the time of slip with a view of reducing long term complications.
Similar differences in approach are seen with regards to prophylactic pinning of the contralateral normal hip joint in patients with unilateral SUFE.
While this is not the norm in the US, it is in German speaking countries. A recent extensive review, weighing up the possible advantages and disadvan- tages now recommends prophylactic pinning also in the US (Schultz et al. 2002).
Whichever general philosophy a surgeon follows, the treatment chosen always has to weigh up the pros and cons for each individual case and guidelines can not be completely prescriptive.
12.8
Imaging Assessment of Surgical
Intervention and Complications of SUFE
The most feared immediate complications of SUFE are chondrolysis and AVN. Chondrolysis is diag- nosed if there is loss of cartilage thickness of ≥ 2 mm compared to the contralateral normal hip or an abso- lute cartilage thickness of ≤ 3 mm in cases of bilater- ally abnormal hips. Cartilage thickness of normal hip joints is assumed to be 4–5 mm (Aronson and Carlson 1992; Loder et al. 1993b; Hughes et al.
1999). The incidence of chondrolysis after SUFE ranges from 1.1%–11.8% (Lubicky 1996; Tudisco et al. 1999). Chondrolysis after SUFE leads to pre- mature closure of the femoral capital growth plate and the apophysis of the greater trochanter. Peri- articular osteopenia on radiographs, bone marrow oedema on MR imaging and generalised increased periarticular activity in bone scintigraphy are seen (Mandell et al. 1992; Warner et al. 1996). Chon- drolysis can be the result of long term immobilisa- tion, direct impingement of the bared metaphysis on the acetabulum or due to surgical intervention with cartilage injury either at the time of surgery or later due to migration of metalwork. An autoim- mune mechanism is also controversially discussed, the exact mechanism leading to chondrolysis is not established yet (Sternlicht et al. 1992; Stover et al. 1994; Lubicky 1996; Warner et al. 1996; Arnold et al. 2002a,b; Exner et al. 2002; Hackenbroch et al. 2002; Leunig et al. 2002; Jofe et al. 2004).
The acute stage of chondrolysis after SUFE lasts about 6 months and presents with joint space nar- rowing. In the longer term 3 outcomes are seen. There
is joint destruction with painful or painless ankylosis or there can be resolution of the pathological find- ings with joint space restoration. The latter outcome is seen in 50%–60% of cases (Lubicky 1996).
AVN presents radiographically with patchy areas of radiolucency and sclerosis and finally epiphyseal collapse (Figs. 12.3, 12.12) (Gekeler 2002). On MR imaging initially oedema like signal is seen within the femoral head. This becomes demarcated by a serpiginous single or less frequently double line, the area enclosed usually demonstrates initially oedema like signal and later fat signal (Kaplan et al. 2001).
The risk is increased in unstable SUFE (Loder et al.
1993b; Ballard and Cosgrove 2002; Tokmakova et al. 2003) and neck of femur or subcapital osteot- omy (Arnold et al. 2002). High grade slips are a risk factor for long-term arthrosis but whether they are also a risk factor for AVN is controversial (Loder et al. 1993b; Arnold et al. 2002; Ballard and Cos- grove 2002; Tokmakova et al. 2003). Active closed reduction is thought to be a risk factor for AVN by many (Aronson and Carlson 1992; Boero et al. 2003; Tokmakova et al. 2003) and for chronic slip this is universally accepted (Boyer et al. 1981;
Aronson and Carlson 1992; Arnold et al. 2002;
Exner et al. 2002; Boero et al. 2003).
Fracture and migration of metal implants, subtro- chanteric femur fractures and infection are cited as further complications of surgical treatment (Hack- enbroch et al. 2002). However, no case of infection after closed reduction has been described in the lit- erature (Schultz et al. 2002).
When assessing operative and postoperative images the following features should be checked (Table 12.3).
What is the postoperative slip angle? Has there been reduction?
Has there been an intraoperative or is there a postoperative chondral injury by surgical implants (risk of chondrolysis)?
If a single cannulated screw is used, is it positioned centrally in the femoral head and is it perpendicular to the growth plate (Aronson and Carlson 1992)?
Are five threads of the screw in the femoral head (the fewer threads the higher the risk of instability and secondary slip; Carney et al. 2003)?
Is there possible impingement of the femoral metaphysis onto the acetabulum?
In longer term follow up (Table 12.4), is there
implant migration? Is there evidence of instability
of a surgically fixed slip? Is there evidence of AVN or
chondrolysis? Is there evidence of secondary degen-
erative change?
12.9
Conclusion and Outlook
Slipped upper femoral epiphysis is not a frequent disease but it is potentially well treatable and early diagnosis is important. In undiagnosed hip, groin, thigh or knee pain, AP and frog lateral radiographs (ideally after Imhäuser) can be diagnostic or reas- sure a normal status.
If SUFE is diagnosed urgent advice and treat- ment from a specialised centre should be sought.
The radiologist’s role is to confirm the diagnosis, help with the classification, advise on imaging in more complex cases and to assess pre- and postop- erative images for possible complications especially mechanical impingement, AVN and chondrolysis.
With the growing obesity problem in the young population SUFE is unfortunately likely to increase in incidence.
References
Adamczyk MJ, Weiner DS et al (2003) A 50-year experience with bone graft epiphysiodesis in the treatment of slipped capital femoral epiphysis. J Pediatr Orthop 23:578–583 Ankarath S, Ng AB et al (2002) Delay in diagnosis of slipped
upper femoral epiphysis. J R Soc Med 95:356–358 Arnold P, Jani L et al (2002a) Results of treating slipped capital
femoral epiphysis by pinning in situ. Orthopade 31:880–
887
Arnold P, Jani L et al (2002b) Management and treatment results for acute slipped capital femoral epiphysis. Ortho- pade 31:866–870
Arnold P, Jani L et al (2002c) Significance and results of sub- capital osteotomy in severe slipped capital femoral epiph- ysis. Orthopade 31:908–913
x AVN or chondrolysis?
x Secondary degenerative change?
situ single screw fixation for stable slipped capital femoral epiphysis. J Pediatr Orthop 23:584–589
Castriota-Scanderbeg A, Orsi E (1993) Slipped capital femo- ral epiphysis: ultrasonographic findings. Skeletal Radiol 22:191–193
Daschner H, Lehner K et al (1990) Imaging of epiphyseolysis of the femur head in the magnetic resonance tomogram.
Rofo Fortschr Geb Rontgenstr Neuen Bildgeb Verfahr 152:583–586
Engelhardt P(2002) Slipped capital femoral epiphysis and the
“healthy” opposite hip. Orthopade 31:888–893
Engelhardt P, Roesler H (1987) Radiometry of epiphyseolysis capitis femoris. A comparison of conventional roentgen study with axial computerized tomography. Z Orthop Ihre Grenzgeb 125:177–182
Exner GU, Schai PA et al (2002) Treatment of acute slips and clinical results in slipped capital femoral epiphysis. Ortho- pade 31:857–865
Freyschmidt J, Brossmann J et al (2001) Slipped capital femoral epiphysis. Borderlands of normal and early pathological findings in skeletal radiography, vol 1. Freyschmidt, Stutt- gart / Thieme, New York, pp 834–836
Gekeler J (2002) Radiology and measurement in adolescent slipped capital femoral epiphysis. Orthopade 31:841–850 Guzzanti V, Falciglia F et al (2003) Slipped capital femoral
epiphysis: physeal histologic features before and after fixation. J Pediatr Orthop 23:571–577
Hackenbroch MH, Kumm DA et al (2002) Dyamic screw fixation for slipped capital femoral epiphysis. Treatment results. Orthopade 31:871–879
Harland U, Krappel FA (2002) Value of ultrasound, CT, and MRI in the diagnosis of slipped capital femoral epiphysis (SCFE). Orthopade 31:851–856
Hughes LO, Aronson J et al (1999) Normal radiographic values for cartilage thickness and physeal angle in the pediatric hip. J Pediatr Orthop 19:443–448
Jofe MH, Lehman W et al (2004) Chondrolysis following slipped capital femoral epiphysis. J Pediatr Orthop B 13:29–31 Kallio PE, Lequesne GW et al (1991) Ultrasonography in
slipped capital femoral epiphysis. Diagnosis and assess- ment of severity. J Bone Joint Surg Br 73:884–889 Kallio PE, Paterson DC et al (1993) Classification in slipped
capital femoral epiphysis. Sonographic assessment of sta- bility and remodeling. Clin Orthop 294:196–203
Kaplan PA, Helms CA et al (2001). Vascular abnormalities of bone. Musculoskeletal MRI, vol 1. Saunders, Philadelphia, pp 334–339
Kelsey JL (1971) The incidence and distribution of slipped capital femoral epiphysis in Connecticut. J Chronic Dis 23:567–578
Kelsey JL, Keggi KJ et al (1970) The incidence and distribu- tion of slipped capital femoral epiphysis in Connecticut and Southwestern United States. J Bone Joint Surg Am 52:1203–1216
Klein A, Joplin RJ et al (1951) Roentgenographic features of slipped capital femoral epiphysis. Am J Roentgenol 66:361–374
Kocher MS, Bishop JA et al (2004) Delay in diagnosis of slipped capital femoral epiphysis. Pediatrics 113:e322–e325 Lalaji A, Umans H et al (2002) MRI features of confirmed
“pre-slip” capital femoral epiphysis: a report of two cases.
Skeletal Radiol 31:362–365
Leunig M, Fraitzl C. R et al (2002) Early damage to the acetabu- lar cartilage in slipped capital femoral epiphysis. Thera- peutic consequences. Orthopade 31:894–899
Loder RT (1996a) The demographics of slipped capital femo- ral epiphysis. An international multicenter study. Clin Orthop 322:8–27
Loder RT (1996b) A worldwide study on the seasonal varia- tion of slipped capital femoral epiphysis. Clin Orthop 322:28–36
Loder RT, Aronson DD et al (1993a) The epidemiology of bilat- eral slipped capital femoral epiphysis. A study of children in Michigan. J Bone Joint Surg Am 75:1141–1147 Loder RT, Richards BS et al (1993b) Acute slipped capital femo-
ral epiphysis: the importance of physeal stability. J Bone Joint Surg Am 75:1134–1140
Loder RT, Mehbod AA et al (2003) Acetabular depth and race in young adults: a potential explanation of the differences in the prevalence of slipped capital femoral epiphysis between different racial groups? J Pediatr Orthop 23:699–702 Lubicky JP (1996) Chondrolysis and avascular necrosis: com-
plications of slipped capital femoral epiphysis. J Pediatr Orthop B 5:162–167
Mandell GA, Keret D et al (1992) Chondrolysis: detection by bone scintigraphy. J Pediatr Orthop 12:80–85
Milz S, Boszczyk A et al (2002) Development and functional structure of the epiphyseal plate. Orthopade 31:835–840 Oppenheim WL, Bowen RE et al (2003) Outcome of slipped
capital femoral epiphysis in renal osteodystrophy. J Pedi- atr Orthop 23:169–174
Poussa M, Schlenzka D et al (2003) Body mass index and slipped capital femoral epiphysis. J Pediatr Orthop B 12:369–371
Schai PA, Exner GU (2002) Indication for and results of inter- trochanteric osteotomy in slipped capital femoral epiphy- sis. Orthopade 31:900–907
Schultz WR, Weinstein JN et al (2002) Prophylactic pinning of the contralateral hip in slipped capital femoral epiphy- sis: evaluation of long-term outcome for the contralateral hip with use of decision analysis. J Bone Joint Surg Am 84A:1305–1314
Smergel EM, Harcke HT et al (1987) Use of bone scintigraphy in the management of slipped capital femoral epiphysis.
Clin Nucl Med 12:349–353
Southwick WO (1967) Osteotomy through the lesser trochan- ter for slipped capital femoral epiphysis. J Bone Joint Surg Am 49:807–835
Stabler A, Genz K et al (1992) MRT of epiphyseolysis capitis femoris. Bildgebung 59:133–135
Sternlicht AL, Ehrlich MG et al (1992) Role of pin protrusion in the etiology of chondrolysis: a surgical model with radio- graphic, histologic, and biochemical analysis. J Pediatr Orthop 12:428–433
Stover B, Sigmund G et al (1994) Early changes in the hip joint following epiphysiolysis of the femoral head. Results of an MRT study. Radiologe 34:46–51
Tokmakova KP, Stanton RP et al (2003) Factors influencing the development of osteonecrosis in patients treated for slipped capital femoral epiphysis. J Bone Joint Surg Am 85A:798–801
Tudisco C, Caterini R et al (1999) Chondrolysis of the hip complicating slipped capital femoral epiphysis: long-term follow-up of nine patients. J Pediatr Orthop B 8:107–111 Umans H, Liebling MS et al (1998) Slipped capital femoral
epiphysis: a physeal lesion diagnosed by MRI, with radio- graphic and CT correlation. Skeletal Radiol 27:139–144 Warner WC Jr, Beaty JH et al (1996) Chondrolysis after slipped
capital femoral epiphysis. J Pediatr Orthop B 5:168–172 Weiner D (1996) Pathogenesis of slipped capital femoral
epiphysis: current concepts. J Pediatr Orthop B 5:67–73
